Information-theoretic analysis of multivariate single-cell signaling responses

Abstract

Mathematical methods of information theory appear to provide a useful language to describe how stimuli are encoded in activities of signaling effectors. Exploring the information-theoretic perspective, however, remains conceptually, experimentally and computationally challenging. Specifically, existing computational tools enable efficient analysis of relatively simple systems, usually with one input and output only. Moreover, their robust and readily applicable implementations are missing. Here, we propose a novel algorithm, SLEMI—statistical learning based estimation of mutual information, to analyze signaling systems with high-dimensional outputs and a large number of input values. Our approach is efficient in terms of computational time as well as sample size needed for accurate estimation. Analysis of the NF- κ B single—cell signaling responses to TNF- α reveals that NF- κ B signaling dynamics improves discrimination of high concentrations of TNF- α with a relatively modest impact on discrimination of low concentrations. Provided R-package allows the approach to be used by computational biologists with only elementary knowledge of information theory.

Fractional response analysis reveals logarithmic cytokine responses in cellular populations

“…The distributions P(y|x j ) can be reconstructed from data using a variety of probability density estimators 43 . The use of the available estimators, however, might be problematic for multivariate responses 24,44 . We therefore propose a more convenient strategy.…”

“…The above strategy allows avoiding estimation of the response distributions, P(y|x j ), from data. For continuous and multivariate variable y the estimation of P(x j |y) is generally simpler than estimation of P(y|x j ) 24,43 . Precisely, an estimator !…”

“…Third, live-cell imaging experiments demonstrated that the dose may not only alter the response at a single time-point but can control temporal profiles of signaling responses 22,23 . For instance, low doses of TNF-α may induce one peak of nuclear factor-κB (NF-κB) signaling activity, whereas higher doses may induce additional peaks 7,24 . Besides, the dose may control the onset, shut off, amplitude, or, in principle, any other characteristics of the responses [25][26][27][28] .…”

Cellular state determines the multimodal signaling response of single cells

“…1a). It therefore remains largely unclear how individual cells are able to accurately integrate the complexity of variable extra-and intracellular information and convert this into a specific, context-aware response with the remarkable and reproducible accuracy necessary to sustain multicellular life.One mechanism is based on temporal integration of the dynamic signaling response, which can provide more information than simple sampling of single time-points 21,26,27 . This, however, requires storage of information, as proposed for the plasma membrane or in the form of slowly accumulating effectors 28,29,30 , which may not facilitate an accurate response at short timescales.…”

“…One mechanism is based on temporal integration of the dynamic signaling response, which can provide more information than simple sampling of single time-points 21,26,27 . This, however, requires storage of information, as proposed for the plasma membrane or in the form of slowly accumulating effectors 28,29,30 , which may not facilitate an accurate response at short timescales.…”